||Algorithms for Routing a Heterogeneous Collection of UAVs
|| Control, Intelligent Systems & Robotics
||Professor Swaroop Darbha, Texas A&M University, College Station, USA
||8 October 2014, Wednesday
||3:00 pm to 4:00 pm
||E5-03-19, Engineering Blk E5, Faculty of Engineering, NUS
| The focus of this talk is the following problem: Given a collection of m UAVs and n gt; m targets to be visited, determine the assignment of targets to UAVs and their sequencing so that (1) every target is on the "tour" of some UAV, (2) some vehicle-target assignment constraints are satisfied and (3) the total distance traveled by the collection is a minimum. This problem is a generalization of the Traveling Salesman Problem (TSP) and differs from TSP in the following ways: (1) the UAVs have motion constraints, (2) the number of UAVs is more than 1 and (3) the collection of UAVs can be heterogeneous owing to their structural design or the payloads (sensors etc) that they carry. The heterogeneity of UAVs can impose vehicle-target assignment constraints; for example, a UAV may not carry the appropriate sensor suite to visit a particular target in which case such an assignment is not useful.
Motion constraints make the problem difficult: the optimal approach conditions of a UAV at every target is a function of the assignment and sequence of targets and vice-versa. As a result of such a strong coupling, one cannot ignore motion constraints in finding the optimal sequence and vice versa. This strong coupling yields conservative lower bounds for the optimal cost and makes the determination of optimal solutions difficult.
In this talk, I will provide an overview of the different kinds of algorithms (approximation algorithms, lower bounding algorithms and heuristics) that our group at TAMU has developed over the last decade and also discuss variants of the routing problem we are currently pursuing.
This is a joint work with my colleague at TAMU, Prof. Siva Rathinam and my colleagues, Phil Chandler and Prof. Meir Pachter at AFRL.
|ABOUT THE SPEAKER|
| Swaroop Darbha received his B. Tech degree in Mechanical Engineering from the Indian Institute of Technology at Madras in 1989, M. S. and Ph. D degrees from the University of California (under the supervision of Prof. Karl Hedrick) at the University of California at Berkeley in 1992 and 1994 respectively. He was a post-doctoral researcher at the California PATH project at UC Berkeley till 1996. Since January 1997, he has been with the Mechanical Engineering Department at Texas A&M University, College Station where he is currently a Professor. He has worked on and contributed to a variety of problems in Intelligent Transportation Systems (platooning algorithms, traffic flow modeling and diagnostic algorithms for air brakes in trucks) and more recently in the area of Unmanned Vehicles (routing and stochastic control). He is currently interested in the broad area of intersection of control systems and combinatorial optimization.
|REMARKS, IF ANY|
| Jointly organized by Edutainment Robotics Lab @ ECE, Social Robotics Lab @ IDMI, Advanced Robotics Center, NUS and Singapore IEEE SMC, R&A and Control Chapters
| Back to seminar list |
All are welcome to attend these Seminars. Should
you have any enquiries, please contact Ms. Lily Png at 6516 6509
or you can e-mail to Ms Lily Png